Cluster headache shows no association with rapid eye movement sleep

Abstract

Background: The connection of cluster headache (CH) attacks with rapid eye movement (REM) sleep has been suggested by various studies, while other authors challenge this assumption. We performed serial polysomnography to determine the association of nocturnal CH attacks and sleep.

Methods: Five patients diagnosed with CH (two with the episodic and three with the chronic subtype) were included and studied over four consecutive nights to evaluate connections between attacks onset and sleep stage.

Results: Twenty typical CH attacks were reported. Thirteen of these attacks arose from sleep. Seven attacks were reported after waking in the morning or shortly before going to sleep. The beginnings of sleep-related attacks were distributed arbitrarily between different non-REM sleep stages. No association of CH attacks with REM or sleep disordered breathing was observed. Increased heart rate temporally associated with transition from one sleep state to another was observed before patients awoke with headache. Total sleep time, total wake time, arousal index and distribution of non-REM sleep stages were different between chronic and episodic CH.

Conclusion: CH attacks are not associated with REM sleep. Brain regions involved in sleep stage transition might be involved in pathophysiology of CH. Differences in sleep characteristics between subgroups might indicate adaptation processes or underlying pathophysiology.

Keywords

Cluster headache trigeminal autonomic cephalalgias polysomnography REM sleep hypothalamus raphe nucleus

Background

A high prevalence of nocturnal sleep-related headache attacks is characteristic for cluster headache (CH). An association of headache onset with rapid eye movement (REM) sleep and sleep disordered breathing (SDB) (1–6) was previously suggested. Recent work questions this causal REM connection (7). Possible REM influence was explained by absent dorsal raphe and locus coeruleus nuclei activity during REM sleep, leading to a reduction of antinociceptive effects of these structures in central pain modulation (8). Hypothalamic pathology was suggested in terms of regulation of circadian rhythm and typical accompanying trigeminal autonomic symptoms (lacrimation, rhinorrhea, tearing, nasal congestion, ptosis, miosis, facial sweating). Imaging data show increased hypothalamic activity during CH attacks (9).

Nine studies report findings from polysomnographic (PSG) monitoring of 104 CH patients (91 with episodic CH and 13 with chronic CH) (1–7,10,11). Some authors report CH attacks to be driven by REM sleep and irregularities of REM sleep (1–4). Others report association with non-REM sleep (5). Headache emerging from REM and non-REM sleep in the same participant with CH was reported by Kudrow and coworkers (5), who monitored ten patients diagnosed with chronic or episodic CH during one night.

Here we report serial PSG findings during four consecutive nights in five patients diagnosed with episodic or chronic CH to determine the temporal association of headache attacks and different sleep stages, to test our hypothesis that neither episodic nor chronic nocturnal CH attacks are associated with REM sleep or SDB. Furthermore, we investigated whether differences in sleep structure and sleep characteristics can be detected comparing CH subtypes.

Patients and Methods

Patients

We investigated a total of five patients with CH (two with the episodic and three with the chronic subtype) fulfilling the ICHD-II diagnostic criteria (code 3.1) (12). All of them were reporting nocturnal CH attacks during similar times almost every night before study inclusion. Each participant was studied during four consecutive nights. Thus, we investigated a total of 20 nights. This number allows reliable results of the temporal association between nocturnal CH attacks and different sleep stages to be gained. No patient received prophylactic medication for CH before or during the study, except one patient diagnosed with chronic subtype, who paused lithium and verapamil 1 week before the first study night. The other two chronic CH patients failed preventive therapy, and patients with the episodic subtype were included before preventive medication was started.

Furthermore, two patients were taking triptans during acute CH attacks on a nearly daily basis. The remaining participants were taking no other medication except for oxygen via nasal probe during acute attacks.

Polysomnography

Each patient underwent overnight PSG monitoring over four consecutive nights using the Embla-System N 7000, Medcare (REMBrandt Manager 7.5) to evaluate the relationship of nocturnal headache attacks to distinct sleep stages and to any occurrence of SDB. PSG was performed during the night time (between 22:00 and 07:00), including two-channel electroencephalography (EEG), electrocardiography (ECG), electro-oculography, chin electromyography (EMG), thoracic and abdominal respiratory efforts measured by impedance plethysmography, body position monitored by a position sensor, oxygen saturation measured by pulse oximetry (ResMedModel 305A, San Diego, CA, USA), surface EMG of tibialis muscles and oronasal airflow recorded by nasal cannula. During the recording patients were monitored by infrared video surveillance. Patients were instructed to sleep as ‘normally’ as possible. Sleep stages were determined visually in according to the standard criteria (13). Arousals were scored according to the criteria of the American Sleep Disorders Association (14).

Apnoea and hypopnoea were defined according to the criteria of the American Sleep Disorders Association. Apnoea was defined as cessation of airflow or reduction of the flow signal <10% of the normal flow for ≥10 s, hypopnoea as an observable reduction of at least 10 s duration followed either by arousal or a desaturation of 4% or more. Events were classified as obstructive or central according to the respiratory effort channels.

The apnoea/hypopnoea index (AHI) was used for quantification of sleep apnoea (number of apnoeas and hypopnoeas per hour of sleep) in accordance to the American Academy of Sleep Medicine Task Force (14). Sleep apnoea was considered mild with AHI between 5 and 15, moderate between 15 and 30 events per hour and severe if AHI was >30. Oxygen indices were calculated by software from the oxygen saturation (SaO₂) curve, with minimal SaO₂ being the lowest saturation reached during sleep and average SaO₂ being the mean of all saturation values reached during all respiratory events. Clear oxygen saturation artifacts were excluded manually before analysis. In case of headache the patients were instructed to inform the attending nurse and record the exact time of awakening and the headache features in a standardized questionnaire.

The protocol of this study was reviewed and approved by the ethics committee of the Faculty of Medicine, University of Duisburg-Essen, and written informed consent was obtained from all participants before study inclusion.

Statistical analysis

Comparison of percentage of CH attacks occurring from different sleep stages and proportion of sleep stages of total sleep time (TST) was performed using the chi-square test. Different characteristics of sleep were compared between subgroups (chronic vs. episodic CH and nights with headache vs. nights without headache) using paired t-test for independent samples. Level of significance was set to p < 0.05. Statistics were calculated with SPSS 18 (SPSS, Inc., Chicago, IL, USA). Results are reported as mean values ± standard deviation (SD).

Results

Individual patient characteristics are shown in Table 1. Summarized PSG results for different CH subgroups are shown in Table 2 for nights with and without headache and for all nights in Table 3. A total of 20 CH headache attacks were reported. Only three patients (one episodic CH, two chronic CH) suffered from nocturnal headache attacks (n = 13). Of these 13 attacks, three (23.0%) arose from non-REM sleep stage 1, nine (69.2%) attacks arose from sleep stage 2 and one attack was reported after wakening from non-REM sleep stage 3 (7.7%). No patient showed headache attacks during REM sleep. Chi-square tests revealed no significant differences of the percentage of the headache attacks arising from the different non-REM sleep stages compared with the percentage of the different sleep stages of TST (CH stage 1 vs. TST stage 1, χ²= 6.0, df = 4, p = 0.199; CH stage 2 vs. TST stage 2, χ²= 6.0, df = 4, p = 0.199; CH stage 3 vs. TST stage 3, χ²= 3.0, df = 2, p = 0.233), suggesting that there is no association between any particular non-REM sleep stage and the onset of CH attacks.

Table 1.

Clinical characteristics of five cluster headache patients

Patient no.	1	2	3	4	5
Gender	M	M	M	M	M
Subtype of CH	Episodic	Episodic	Chronic	Chronic	Chronic
Age at study inclusion (years)	53	51	60	50	58
Age at disease onset (years)	42	41	49	42	47
Usual medication Prophylaxis	–	–	–	Lithium, verapamil	–
During attacks	Oxygen	Oxygen	Oxygen	Triptans	Triptans
BMI (kg/m²)	27.2	29.8	27.8	34.1	24.1
Location of headache	Left	Right	Right	Right	Left
Pain quality	Stabbing	Stabbing	Stabbing	Stabbing	Stabbing
Headache intensity (NRS)	8	7	5	5–6	7
Nausea	–	Occasionally	–	–	–
Vomiting	–	–	–	–	–
Phonophobia	–	–	–	–	–
Photophobia	–	–	–	–	–
Ipsilateral rhinorrhoea	+	+	Occasionally	Occasionally	+
Ipsilateral tearing	+	+	Occasionally	Occasionally	+
Ipsilateral sweating	+	+	Occasionally	Occasionally	+
Ipsilateral eye redness	+	+	+	+	–

BMI: body mass index, CH: cluster headache, M: male, NRS: numeric rating scale (0: no pain, 10: worst imaginable pain); –: absent, + : present.

Table 2.

Sleep characteristics of nights with headache vs. nights without headache

	Nights with headache mean ± SD [range]		Nights without headache mean ± SD [range]
	eCH	cCH	eCH	cCH
Total sleep time (TST) (min)	253 ± 46	349 ± 37	244 ± 97*^‡	297 ± 49
Total sleep time (TST) (min)	[186–288]	[331–423]	[137–348]	[202–340]
Total wake time (TWT) (min)	64 ± 27	48 ± 10	115 ± 88*^‡	81 ± 26^†
Total wake time (TWT) (min)	[45–111]	[38–62]	[27–207]	[53–113]
Sleep latency (min)	32 ± 35	14 ± 9*	29 ± 13	39 ± 31
Sleep latency (min)	[2–82]	[4–27]	[13–99]	[5–36]
Sleep stage measures, as a percentage of TST
Stage 1	15 ± 5	8 ± 3	13 ± 10*	7 ± 6
Stage 1	[9–21]	[5–12]	[5–24]	[3–11]
Stage 2	53 ± 7	66 ± 6	60 ± 18*	67 ± 12
Stage 2	[42–59]	[58–73]	[43–76]	[59–70]
Stage 3	17 ± 4	7 ± 3	8 ± 7	10 ± 6^†
Stage 3	[12–20]	[3–12]	[1–16]	[3–16]
Stage 4	4 ± 4	3 ± 3	8 ± 10	1 ± 3
Stage 4	[0–9]	[0–7]	[0–22]	[0–6]
REM	11 ± 8	16 ± 6	10 ± 12	15 ± 9
REM	[0–19]	[9–26]	[0–23]	[0–25]
AHI	9 ± 5	20 ± 31	11 ± 4	5 ± 7
AHI	[5–17]	[1–80]	[5–15]	[1–18]
Minimal SaO₂ (%)	87 ± 3	88 ± 8	84 ± 5	92 ± 3
Minimal SaO₂ (%)	[82–89]	[72–93]	[83–89]	[88–95]
Arousal Index	26 ± 8	35 ± 18	36 ± 19*^‡	17 ± 6
Arousal Index	[18–35]	[20–60]	[20–57]	[11–27]
Sleep efficiency (%)	70 ± 13	83 ± 5	63 ± 25*^‡	68 ± 13^†
Sleep efficiency (%)	[53–84]	[75–88]	[36–87]	[48–78]

AHI: apnoea/hypopnoea index, cCH: chronic cluster headache, eCH: episodic cluster headache, REM: rapid eye movement, SaO₂: oxygen saturation SD: standard deviation. *p < 0.05 vs. nights with headache in eCH, ^†p < 0.05 vs. nights with headache in cCH, ^‡p < 0.05 vs. nights without headache in cCH.

Table 3.

Sleep characteristics of patients with episodic vs. chronic cluster headache, during all nights

	Episodic CH mean ± SD [range]	Chronic CH mean ± SD [range]
Total sleep time (TST) (min)	248 ± 70 [137–348]	323 ± 50 [202–423]
Total wake time (TWT) (min)	90 ± 66 [27–207]	64 ± 25 [38–113]*
Sleep latency (min)	27 ± 25 [2–99]	26 ± 26 [4–36]
Sleep stage measures, as a percentage of TST
Stage 1	14 ± 7 [5–21]	8 ± 5 [3–12]
Stage 2	56 ± 13 [42–76]	66 ± 9 [58–73]
Stage 3	13 ± 7 [1–20]	9 ± 5 [3–16]
Stage 4	6 ± 7 [0–22]	2 ± 3 [0–7]*
REM	10 ± 10 [0–23]	15 ± 7 [0–26]
AHI	10 ± 4 [5–17]	12 ± 23 [1–80]
Minimal SaO₂ (%)	85 ± 4 [82–89]	90 ± 6 [72–95]
Arousal index	31 ± 14 [18–57]	26 ± 16 [11–60]
Sleep efficiency (%)	67 ± 19 [36–87]	76 ± 12 [48–88]

AHI: apnoea/hypopnoea index, cCH: chronic cluster headache, eCH: episodic cluster headache, REM: rapid eye movement, SaO₂: oxygen saturation, SD: standard deviation. *p < 0.05 vs. night with headache.

None of the investigated patients showed attacks exclusively associated to the same particular non-REM sleep stage across all four investigated nights. Headache occurred from two different non-REM sleep stages within the same night during two study nights in one patient. Serial PSG of one example patient is shown in Figure 1.

Figure 1.

Serial polysomnography of one study patient (patient 1, as an example), showing CH attacks not associated with REM sleep over four consecutive nights (A–D). A clear association with a distinct sleep stage cannot be observed. Arrows indicate the wake-up time of this patient with typical headache. Sleep stages are indicated on the right (REM, rapid eye movement sleep; 1–4, non-REM sleep stages 1–4; W, waking; MT, movement; N, not in bed). Time is indicated on horizontal bars below each hypnogram.

The percentage of total wake time (TWT) was higher (97 ± 60 vs. 61 ± 19 min; p = 0.004), and stage 1 non-REM sleep (10 ± 8 vs. 12 ± 5% of TST; p = 0.019) and sleep efficiency (65 ± 19 vs. 75 ± 10%; p = 0.007) were lower in nights without headache compared with nights with headache in all patients. With respect to subtypes of CH (episodic vs. chronic), not only TWT (p = 0.02) but also amount of non-REM stage 3 sleep (as a percentage of TST; p = 0.027) was higher, whereas sleep efficiency was lower (p = 0.003), during nights without headache than nights with headache in patients with chronic CH. Participants with episodic CH showed higher values for TWT (p = 0.003), TST (p = 0.024) and amount of stage 2 non-REM sleep (as a percentage of TST; p = 0.004) as well as arousal index (p = 0.01), and stage 1 non-REM sleep (as a percentage of TST; p = 0.024) and sleep efficiency (p = 0.037) were lower during nights without headache than nights with headache.

Comparing participants with chronic CH with patients with episodic CH, TWT (p = 0.001) and arousal index (p = 0.004) were lower in chronic CH, whereas TST (p = 0.04) and sleep efficiency (p = 0.017) were lower in patients with episodic CH during nights without headache. Furthermore, sleep latency was lower (p = 0.045) in patients with chronic vs. episodic CH during nights with headache attacks (Table 2).

Comparing sleep characteristics during all nights in patients with episodic CH to patients with chronic CH (Table 3), TWT (p = 0.005) and non-REM stage 4 sleep (as a percentage of TST; p = 0.031) were lower in patients with chronic CH, whereas sleep efficiency tended to be higher in cCH (p = 0.06). Other PSG parameters did not show any differences.

Wake-up times due to headache and corresponding sleep stages and corresponding AHI indices are summarized in Table 4. Three patients showed signs of SDB (AHI > 5 during more than one night), but mean AHI was not different in nights with headache from nights without headache (p = 0.169), and no temporal relationship of SDB with the onset of headache could be established in any of the patients.

Table 4.

Association of sleep stage and time of headache occurrence in patients with cluster headache

Patient	Age (years)	No. of night	No. of headache attacks	Sleep stage before headache awakening	Time	AHI
1	53	1	1	non-REM 1	01:45	7.1
		1	2	non-REM 2	03:37
		2	1	non-REM 2	23:40	5.2
		2	2	non-REM 3	01:42
		3	1	non-REM 2	23:55	16.5
		4	1	non-REM 2	02:56	8.7
4	50	1	1	non-REM2	02:43	80.4
		3	1	non-REM2	06:00	23.4
		4	1	non-REM2	02:35	5.1
			2	non-REM2	06:30
5	58	2	1	non-REM 1	23:19	6.2
		3	1	non-REM 2	23:35	1.0
		4	1	non-REM 1	23:40	2.1

AHI: apnoea/hypopnoea index, REM: rapid eye movement.

Discussion

Our data show no REM association of nocturnal CH attacks in patients suffering from episodic or chronic CH. We recorded 20 typical CH attacks in five patients each monitored by PSG during four consecutive nights. Seven attacks occurred from wakefulness and 13 attacks arose from sleep. All 13 attacks were reported after awakening from non-REM sleep (predominantly sleep stage 2). These findings contradict previously published reports (4) suggesting a strict REM relationship and support studies that report no association of nocturnal headache attacks with REM sleep in patients with CH (5,7). Different distribution of onset of nocturnal headache attacks in REM sleep and non-REM sleep between patients with episodic CH and patients with chronic CH was reported previously (5). The authors suggested that episodic CH attacks more likely occur from REM sleep, whereas chronic CH attacks may also occur from non-REM sleep (5). Our data do not support this distinction, as neither episodic nor chronic CH patients had headache attacks in REM sleep.

Our observation challenges the previously suggested pathophysiological concept of hypothalamic alteration in CH and favours malfunction of other brain stem nuclei associated with sleep control and pain.

Neurons of dorsal raphe nucleus (DRN) have been hypothesized to have a key role in primary headache and were shown to be highly active during headache attacks in patients suffering from, for example, migraine according to recent imaging studies (15). Serotoninergic neurons of this brain area have also shown to have a major role in the regulation of wakefulness and sleep (16). Our participants showed more stage 2 sleep with reduced amount of slow wave sleep (non-REM stage 3 and stage 4) and lower amount of REM sleep (as a percentage of TST) than healthy older participants reported by other authors (17,18). The decreased amount of REM sleep might be an indicator of increased activity of the DRN, which was shown to inhibit REM sleep via serotoninergic innervations. The DRN has been shown to have such efferent interactions to the cerebral cortex, amygdala, basal forebrain, thalamus, locus coeruleus, pontine reticular formation and pre-optic and hypothalamic areas (16).

Whether the difference in the amount of the different sleep stages is secondary to the recurring nocturnal headache attacks, or whether it is a sign of disturbed sleep regulating pathways due to CH pathogenesis, remains unclear. However, this difference in sleep might only be secondary to increased amount of wake after sleep onset due to nocturnal headache attacks. In contrast, the sleep architecture during nights with headache compared with nights without headache did not differ substantially (Table 2).

Three patients (60%) showed at least some signs of SDB and its main subtype obstructive sleep apnoea (OSA) (AHI > 5), a larger proportion than reported in healthy middle aged men (19). This might be due to the number of participants or the higher age of our participants. Nevertheless, no temporal correlation of headache onset and decreased SaO₂ was detected in any of the participants. None of the patients complained of clinical symptoms of OSA (hyper somnolence, fatigue, impaired concentration or increased daytime sleepiness). Moreover, headache attacks also occurred under adequate positive airway pressure (CPAP) therapy (oxygen via nasal probe). These findings indicate no strong association of nocturnal CH attacks and SDB.

Comparing sleep characteristics of the different CH subtypes, we found a highly significant (p = 0.001) difference in TWT, with higher values for patients with episodic CH during nights without headache. This effect might be due to adaptation to regular nocturnal headache, which might be larger in chronic CH. This hypothesis would be in line with the significantly lower arousal index in patients with chronic CH during nights without headache and a trend towards higher values for sleep efficiency in this subgroup (77% vs. 67% in episodic CH, not significant) and lower sleep latency in chronic CH patients.

A major problem in studying CH attacks in regard to headache onset using PSG has to be addressed. It is not possible to determine the exact time of headache onset because the headache generally starts during sleep. Only the wake-up time of the patient with already existing headache can be identified accurately. Some authors reported an increase in heart rate after nociceptive stimuli during sleep (20). We hypothesize a similar reaction to headache in our CH patients, because a rise in heart rate was observed shortly before waking up with a headache attack. This heart rate increase started approximately 7.2 ± 5 min before waking up with the typical CH attack. Thus this might represent the accurate time of headache onset. If this is the case, onset of headache was loosely associated with sleep stage transitions within 6.5 ± 4 min. Transitions were made from non-REM stage 1 to stage 2 in six cases, from stage 2 to stage 3 non-REM sleep in three cases and from stage 2 to stage 1 in one case. This finding might indicate a connection of the initiation of nocturnal CH attacks with neuronal control of transition to a different sleep stage.

Which brain areas are involved in non-REM sleep stage transitions is still unclear. Recent literature indicates the following areas to be involved in regulation of sleep and wakefulness: cholinergic neurons of the laterodorsal tegmental nuclei (LDT) and pedunculopontine tegmental nuclei; monaminergic cells of locus coeruleus and the serotoninergic median and DRN; and glutamatergic neurons in the parabrachial nucleus and the adjacent precoeruleus area and more rostral brain areas (see Saper et al. (21) for a detailed review). These areas are most likely to be involved in regulation of transitions from wakefulness to non-REM and REM sleep. Modified or maladaptive interactions between these different brainstem structures might be involved in the regulation of transitions between non-REM sleep stages. For example cholinergic neurons of the LDT have been shown to increase their firing rates just before the transition from cortical slow waves to faster EEG frequencies in rat (22).

Our study is limited by the small number of patients studied, but to our knowledge results of four consecutive nights in CH patients have not been reported previously. This leads to a total of 13 nocturnal and seven daytime attacks, and thus we have gained insight into the temporal association of CH attacks and different sleep stages.

In conclusion, our data indicate that REM sleep and SDB are not associated with the occurrence of nocturnal headache attacks in patients with episodic or chronic CH. The identified connection to sleep stage transitions as trigger factor for the individual headache attack might be of interest for further research, and may indicate a major role of brainstem structures in nocturnal CH attacks, such as the DRN or LDT nuclei. Different sleep architecture in episodic and chronic CH indicates diverse pathophysiological mechanisms in both subtypes.

Footnotes

Author contributions

SZ was involved in designing the study protocol and the data analysis, conducted the statistical analysis and wrote the manuscript.

DH was involved in designing the study protocol and helped in performing the experiments and preparing the article.

TEW analysed the PSG data and contributed to the final manuscript.

HCD helped prepare the manuscript.

SK participated in designing the study protocol and writing the article.

MO designed the study protocol and participated in analysis of the data and in the writing process of the manuscript.

Declaration of competing interests

DH has received a scientific grant from Grünenthal not related to this study.

HCD has received honoraria for participation in clinical trials, contribution to advisory boards or lectures from Addex Pharma, Allergan, Almirall, AstraZeneca, Bayer Vital, Berlin Chemie, Coherex Medical, CoLucid, Böhringer Ingelheim, Bristol-Myers Squibb, GlaxoSmithKline, Grünenthal, Janssen-Cilag, Lilly, La Roche, 3M Medica, Minster, MSD, Novartis, Johnson & Johnson, Pierre Fabre, Pfizer, Schaper and Brümmer, SanofiAventis and Weber & Weber; has received research support from Allergan, Almirall, AstraZeneca, Bayer, GlaxoSmithKline, Janssen-Cilag and Pfizer. The financial support is not related to this study.

ZK has received research grants and honoraria from Allergan, Bayer, Biogen and Merck and is an advisory board member for Allergan, not related to this study.

MO has received scientific support and/or honoraria from Biogen Idec, Novartis, Sanofi-Aventis, Pfizer and Teva. He received research grants from the German ministry for education and research (BMBF), not related to this study.

SK, SZ and TEW report no disclosures.

References

Nobre

Leal

Filho

. Investigation into sleep disturbance of patients suffering from cluster headache. Cephalalgia 2005; 25: 488–492.

Nobre

Filho

Dominici

. Cluster headache associated with sleep apnoea. Cephalalgia 2003; 23: 276–279.

Della Marca

Vollono

Rubino

Capuano

Di Trapani

Mariotti

. A sleep study in cluster headache. Cephalalgia 2006; 26: 290–294.

Buckle

Kerr

Kryger

. Nocturnal cluster headache associated with sleep apnea. A case report. Sleep 1993; 16: 487–489.

Kudrow

McGinty

Phillips

Stevenson

. Sleep apnea in cluster headache. Cephalalgia 1984; 4: 33–38.

Chervin

Zallek

Lin

Hall

Sharma

Hedger

. Sleep disordered breathing in patients with cluster headache. Neurology 2000; 54: 2302–2306.

Terzaghi

Ghiotto

Sances

Rustioni

Nappi

Manni

. Episodic cluster headache: NREM prevalence of nocturnal attacks. Time to look beyond macrostructural analysis? Headache 2010; 50: 1050–1054.

Lydic

McCarley

Hobson

. Serotonin neurons and sleep. II. Time course of dorsal raphe discharge, PGO waves, and behavioral states. Arch Ital Biol 1987; 126: 1–28.

Morelli

Pesaresi

Cafforio

. Functional magnetic resonance imaging in episodic cluster headache. J Headache Pain 2009; 10: 11–14.

10.

Vetrugno R, Pierangeli G, Leone M, Bussone G, Franzini A, Brogli G, D'Angelo R, Cortelli P, and Montagna P. Effect on sleep of posterior hypothalamus stimulation in cluster headache. Headache Jul-Aug 2007; 47(7): 1085–90.

11.

Khatami R, Tartarotti S, Siccoli MM, Bassetti CL, Sándor PS. Long-term efficacy of sodium oxybate in 4 patients with chronic cluster headache. Neurology 2011 Jul 5; 77(1): 67–70. Epub 2011 May 25.

12.

InternationalHeadacheSociety. The International Classification of Headache Disorders: 2nd edition. Cephalalgia 2004; 24(Suppl 1): 9–160.

13.

Hori

Sugita

Koga

. Proposed supplements and amendments to ‘A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects', the Rechtschaffen & Kales (1968) standard. Psychiatry Clin Neurosci 2001; 55: 305–310.

14.

American Sleep Disorders Association. EEG arousals: scoring rules and examples: a preliminary report from the Sleep Disorders Atlas Task Force of the American Sleep Disorders Association. Sleep 1992; 15: 173–184.

15.

Demarquay

Lothe

Royet

. Brainstem changes in 5-HT1A receptor availability during migraine attack. Cephalalgia 2011; 31: 84–94.

16.

Monti

. Serotonin control of sleep-wake behavior. Sleep Med Rev 2011; 15: 269–281.

17.

Landolt

Dijk

Achermann

Borbely

. Effect of age on the sleep EEG: slow-wave activity and spindle frequency activity in young and middle-aged men. Brain Res 1996; 738: 205–212.

18.

Vitiello

Larsen

Moe

Borson

Schwartz

Prinz

. Objective sleep quality of healthy older men and women is differentially disrupted by nighttime periodic blood sampling via indwelling catheter. Sleep 1996; 19: 304–311.

19.

Young

Palta

Dempsey

Skatrud

Weber

Badr

. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993; 328: 1230–1235.

20.

Lavigne

Zucconi

Castronovo

. Heart rate changes during sleep in response to experimental thermal (nociceptive) stimulations in healthy subjects. Clin Neurophysiol 2001; 112: 532–535.

21.

Saper

Fuller

Pedersen

Scammell

. Sleep state switching. Neuron 2010; 68: 1023–1042.

22.

Boucetta

Jones

. Activity profiles of cholinergic and intermingled GABAergic and putative glutamatergic neurons in the pontomesencephalic tegmentum of urethane-anesthetized rats. J Neurosci 2009; 29: 4664–4674.